Scribd
Upload
4 views2 pages

Exponential Smoothing in Forecasting

The document provides an introduction to Exponential Smoothing, a technique used for forecasting time-series data, particularly in inventory management and financial forecasting. It explains the Simple Exponential Smoothing (SES) method, detailing the formula used for forecasting and the significance of the smoothing constant. An example calculation demonstrates how to apply the method to forecast sales, concluding with a forecast of 218.56 units for Period 4.

Uploaded by

reiddell12
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
4 views2 pages

Exponential Smoothing in Forecasting

The document provides an introduction to Exponential Smoothing, a technique used for forecasting time-series data, particularly in inventory management and financial forecasting. It explains the Simple Exponential Smoothing (SES) method, detailing the formula used for forecasting and the significance of the smoothing constant. An example calculation demonstrates how to apply the method to forecast sales, concluding with a forecast of 218.56 units for Period 4.

Uploaded by

reiddell12
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

ST. MARK COLLEGE of baliuag, Bulacan, inc.

JP. Rizal St. Sta. Barbara, Baliuag Bulacan


Email: stmark04@[Link] | Tel. No. 305-5264
Name: _______________________________ Date: _________________
Subject: Business Mathematics/Forecasting
Module #: 11

Forecasting: Introduction to Exponential Smoothing


I. Introduction to Exponential Smoothing
Exponential Smoothing is a rule-of-thumb technique for smoothing time-series data, often
used in inventory management and financial forecasting. It produces a forecast that is
essentially a weighted average of past observations, with the weights decaying exponentially
as the observations get older. In simpler terms, this means recent data is given more
importance than older data when calculating the next forecast.

It is particularly useful for short-term forecasts (like the next period's demand) where the
underlying data has no significant trend or seasonal patterns, as it is simple to compute and
requires minimal historical data.

II. Simple Exponential Smoothing (SES)


The most basic form of this method is Simple Exponential Smoothing (SES). This method is
applied when the data series is stationary, meaning the average value (the mean) does not
change over time.

The Formula
The forecast for the next period, , is calculated using the following equation:

Where:
● = The new forecast for the next period ().
● = The actual demand observed in the current period ().
● = The previous forecast made for the current period ().
● (alpha) = The smoothing constant (a value between and ).
The formula shows that the new forecast is a blend of the most recent actual data () and the
previous forecast ().

The Smoothing Constant ()


The smoothing constant, , determines how much weight is given to the most recent actual
demand versus the prior forecast.
1. If is close to 1 (e.g., ): The forecast is highly sensitive to recent actual demand. This is
best used for data that is unstable or has rapid shifts, as the new forecast will quickly
adjust to new information. The old forecast () is heavily discounted.
2. If is close to 0 (e.g., ): The forecast gives very little weight to the most recent actual
demand and relies heavily on the smoothed average (the previous forecast, ). This is
best used for stable data where any single large fluctuation is considered a random
error, not a true change in the pattern.
A typical starting value for is often . The best value is determined by minimizing the forecast
error over time.

III. Example Calculation


Imagine a store wants to forecast sales for Period 4. They choose a smoothing constant .

Period (t) Actual Previous Calculation New Forecast


Demand (Dt) Forecast (Ft) for Ft+1 (Ft+1)

1 220 200 (Initial 204


Guess)

2 250 204 213.2

3 240 213.2 218.56

Conclusion: The forecast for Period 4 is 218.56 units.

Note: For the very first period (Period 1), the previous forecast (or ) must be an initial guess,
such as the average of past demand or the most recent demand.

You might also like